摘要:
In apparatuses for automatically acquiring and also for automatically classifying images of defects present on a sample such as a semiconductor wafer, a classifying system is provided which are capable of readily accepting even such a case that a large number of classification classes are produced based upon a request issued by a user, and also even such a case that a basis of the classification class is changed in a high frequency. When the user defines the classification classes, a device for designating attributes owned by the respective classification classes is provided. The classifying system automatically changes a connecting mode between an internally-provided rule-based classifier and an example-based classifier, so that such a classifying system which is fitted to the classification basis of the user is automatically constructed.
摘要:
Provided is an examination technique to detect defects with high sensitivity at an outer-most repeative portion of a memory mat of a semiconductor device and even in a peripheral circuit having no repetitiveness.A circuit pattern inspection apparatus comprises an image detection unit for acquiring an image of a circuit pattern composed of multiple die having a repeative pattern, a defect judgement unit which composes, in respect of an acquired detected image, reference images by switching addition objectives depending on regions of repeative pattern and the other regions and compares a composed reference image with the detected image to detect a defect, and a display unit for displaying the image of the detected defect.
摘要:
A classification model optimum for realization of a defect classification request by a user is not known by the user. Then, the user sets a classification model which is not necessarily suitable and makes classification, resulting in degradation in classification performance. Therefore, the present invention automatically generates plural potential classification models and combines class likelihoods calculated from the plural classification models to classify. To combine, an index about the adequacy of each model, in other words, an index indicating a reliable level of likelihood calculated from the each potential classification model, is also calculated. Considering the calculated result, the class likelihoods calculated from the plural classification models are combined to execute classification.
摘要:
A method and apparatus for detecting defects includes irradiating and scanning an electron beam focused on an area of a sample, detecting charged particles generated from the sample by the irradiating and scanning of the electron beam with a first detector which detects charged particles having relatively low energy to obtain a first image of the area and with a second detector which detects charged particles having relatively high energy to obtain a second image of the area, comparing the first inspection image of the area with a first reference image to generate a first difference image, and comparing obtained second image of the area with a second reference image to generate a second difference image, and detecting an open defect or a short defect from at least one of the generated first difference image and the second difference image.
摘要:
A defect review method and device of the invention solves the previous problem of a long inspection time that is caused by the increase of a process-margin-narrow pattern as a result of the size reduction of a semiconductor device. With the method and device of the invention, an SEM (Scanning Electron Microscope) image is derived by capturing an image of a process-margin-narrow pattern portion extracted based on lithography simulation with image-capturing conditions of a relatively low resolution. The resulting SEM image is compared with CAD (Computer Aided Design) data for extraction of any abnormal section. An image of the area extracted as being abnormal is captured again, and the resulting high-resolution SEM image is compared again with the CAD data for defect classification based on the feature amount of the image, e.g., shape deformation. The abnormal section is then measured in dimension at a position preset for the classification result so that the time taken for inspection can be prevented from increasing.
摘要:
A hotspot searching apparatus manufactures a small number of chips or regions on a semiconductor wafer under respectively different manufacturing process conditions, compares SEM images of their external appearances to output a point having large differences as a narrow process window, that is, a process monitoring point that should be managed in mass production, the narrow process window having a narrow manufacturing process condition (exposure condition) in the manufacturing of the semiconductor wafer, and sets the point as a measurement point by a CD-SEM apparatus, such that it extracts and determines plural circuit pattern parts having a narrow manufacturing process margin as the process monitoring point in a short time and a process monitoring point monitoring performs shape inspection or shape length measurement in detail at high resolution.
摘要:
Defect image display screens are capable of accurately presenting features of defects. On a thumbnail display screen of a defect, images likely to most clearly indicating features of the defect are determined in units of the defect from, for example, inspection information and a defect type, and then are displayed. On a detail display screen of a defect, for example, images for being displayed so as to clearly indicate features of the defect, and the display sequence thereof are determined in accordance with, for example, inspection information and a defect type, and then are displayed. Further, steps for acquiring a display image during or after defect image acquisition by using, for example, a different defect image acquisition apparatus and a different imaging condition in accordance with preliminarily specified rules are added to an imaging sequence (procedure).
摘要:
A method of observing a specimen using a scanning electron microscope, makes it possible to shorten the time required to perform automatic focusing at the time of semiconductor defect automatic review and improves the throughput in the processing in which the specimen is observed. In the above method, the specimen is imaged at a low resolution by the scanning electron microscope to obtain an image, an area for imaging the specimen at a high resolution is specified from the image acquired at the low resolution, the specimen is imaged at a high resolution by the scanning electron microscope to determine a focus position, a focal point of the scanning electron microscope is set to the determined focus position, and a high resolution image in the specified area is acquired in a state in which the focus position has been set to the determined focus position.
摘要:
An apparatus and a method for automatically inspecting a defect by an electron beam using an X-ray detector. The composition of a defective portion is analyzed with higher rapidity and the cause of the defect is easily and accurately determined based on an X-ray spectrum. The X-ray spectrum and the image of foreign particles formed on a process QC wafer are registered as reference data, and the defects generated on a process wafer are classified by collation with the reference data. The use of both the X-ray spectrum and the detected image optimizes the operating conditions for X-ray detection. A defect of which the X ray is to be detected is selected based on the result of classification of defect images automatically collected, and the defect is classified according to the features including both the composition and the external appearance.
摘要:
In performing a programmed-point inspection of a circuit pattern using a review SEM, stable inspection can be performed while suppressing the generation of a false report even when a variation in a circuit pattern to be inspected is large. SEM images that are obtained by sequentially imaging a predetermined circuit pattern using the review SEM are stored into a storage unit. Images that meet a set condition are selected from the stored SEM images, and averaged to create an average image (GP image). By performing pattern check by GP comparison using this GP image, an inspection can be performed while suppressing the generation of a false report even when a variation in the circuit patterns is large.